CLJun 29, 2017
Frame-Based Continuous Lexical Semantics through Exponential Family Tensor Factorization and Semantic Proto-Roles
arXiv:1706.09562v11091 citations
Originality Incremental advance
AI Analysis
This work addresses lexical semantics for natural language processing, presenting an incremental improvement.
The paper tackled the problem of learning continuous lexical semantics by integrating different frame annotations, achieving multiple 10% gains over baselines in encoding syntactic-semantic content.
We study how different frame annotations complement one another when learning continuous lexical semantics. We learn the representations from a tensorized skip-gram model that consistently encodes syntactic-semantic content better, with multiple 10% gains over baselines.